Using the Blinder-Oaxaca Decomposition Method to Measure Racial Bias in Achievement Tests

The Blinder-Oaxaca decomposition technique is typically applied to gender or racial earnings gaps with the goal of determining the percent of the gap that can be attributed to differences in attributes between groups and to labor market discrimination. We apply this technique to the racial gap in achievement tests with the goal of measuring the relative racial bias of these tests. The Armed Services Vocational Aptitude Battery (ASVAB) and the Peabody Individual Achievement Test (PIAT) were administered to respondents from the 1997 National Longitudinal Survey of Youth. Decomposition results indicate that up to 49.5 % of the white-black gap in the PIAT is explained by racial differences in average attributes. The corresponding figure for the ASVAB is 44.7 %. Since the same sample and specification are used in estimating ASVAB and PIAT scores, the difference in the percent of the gaps due to attributes can be ascribed to test score bias and not omitted variable bias. Therefore, the results suggest that the “discrimination” or bias of the ASVAB is greater than the PIAT. Copyright Springer Science+Business Media New York 2013